SIST EN ISO 1886:1999

SLOVENSKI STANDARD
SIST EN ISO 1886:1999
01-maj-1999
9ODNQD]DRMDþHQMH±8SRUDEQHPHWRGHY]RUþHQMD,62
Reinforcement fibres - Sampling plans applicable to received batches (ISO 1886:1990)
Verstärkungsfasern - Stichprobenanweisungen für die Loseingangsprüfung (ISO
1886:1990)
Fibres de renfort - Méthodes d'échantillonnage pour le contrôle de réception de lots (ISO
1886:1990)
Ta slovenski standard je istoveten z: EN ISO 1886:1994
ICS:
59.100.01 0DWHULDOL]DRMDþLWHY Materials for the
NRPSR]LWRYQDVSORãQR reinforcement of composites
in general
SIST EN ISO 1886:1999 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

---------------------- Page: 1 ----------------------

SIST EN ISO 1886:1999

---------------------- Page: 2 ----------------------

SIST EN ISO 1886:1999

---------------------- Page: 3 ----------------------

SIST EN ISO 1886:1999

---------------------- Page: 4 ----------------------

SIST EN ISO 1886:1999
- INTERNATIONAL
STANDARD
Third edition
1990-12-15
Reinforcement fibres - Sampling plans
applicable to received batches
Fibres de renfort
- Mgthodes d ’khantillonnage pour le contrdle de
rkep tion de lots
Reference number
IS0 1886: 1990(E)

---------------------- Page: 5 ----------------------

SIST EN ISO 1886:1999
IS0 1886:1990(E)
Foreword
lS0 (the International Organization for Standardization) is a worldwide
federation of national standards bodies (IS0 member bodies). The work
of preparing International Standards is normally carried out through IS0
technical committees. Each member body interested in a subject for
which a technical committee has been established has the right to be
International organizations, govern-
represented on that committee.
mental and non-governmental, in liaison with ISO, also take part in the
work. IS0 collaborates closely with the International Electrotechnical
Commission (IEC) on all matters of electrotechnical standardization.
Draft International Standards adopted by the technical committees are
circulated to the member bodies for voting. Publication as an lnter-
national Standard requires approval by at least 75 % of the member
bodies casting a vote.
International Standard IS0 1886 was prepared by Technical Committee
ISO/TC 61, Plastics.
This third edition cancels and replaces the second edition
(IS0 1886:1980), of which it constitutes a technical revision.
0 IS0 1990
All rights reserved. No part of this publication may be reproduced or utilized in any form
or by any means, electronic or mechanical, including photocopying and microfilm, without
permission in writing from the publisher.
International Organization for Standardization
Case Postale 56 l Cl-l-1211 Geneve 20 l Switzerland
Printed in Switzerland
ii

---------------------- Page: 6 ----------------------

SIST EN ISO 1886:1999
IS0 1886:1990(E)
Introduction
As indicated in the title, this International Standard is designed for re-
ceived batches and not for inspection during manufacturing of a product.
In the latter case: the manufacturer should have the necessary facilities
available and a good knowledge of the product, thus allowing him to
assure the quality through an appropriate control system.
For checking received batches, the customer has only limited infor-
mation about a given product.
The sampling plans described hereafter are intended as “normal” plans,
and require a certain number of results for the assessment of a batch
to a given confidence level.
Other sampling plans, i.e. reduced or tightened plans, can be con-
sidered, depending on the product type, the application for which the
product is intended, the test method and the degree of inspection re-
quired.
The choice of sampling plan and the extent of sampling depends on
-
the knowledge of the product acquired by the customer during prod-
uct qualification, plus information from routine inspections of re-
ceived batches;
- th e degree of confident e that the customer is prepared to accord the
undertaken b y the manufacturer.
in spections
. . .
Ill

---------------------- Page: 7 ----------------------

SIST EN ISO 1886:1999
This page intentionally left blank

---------------------- Page: 8 ----------------------

SIST EN ISO 1886:1999
INTERNATIONAL STANDARD IS0 18863 990(E)
- Sampling plans applicable to received
Reinforcement fi bres
batches
some of which are versions of definitions already
1 Scope
published in IS0 3534, for instance, but reworded to
suit the particular context of this International Stan-
This International Standard specifies two methods
dard.
of sampling - by attributes or by measurements
(variables) - applicable to batches of reinforcement
3.1 sampling plan: A series of rules established to
materials (in particular textile glass, carbon fibre
obtain, through the selection of a sample and its
and aramid fibre) in various forms (e.g. package,
subsequent analysis, an evaluation which is as re-
roll, bulk material). For both methods, this Inter-
liable as possible of the quality of a received batch,
national Standard includes tables with acceptance
so as to determine, using criteria defined in the
and rejection criteria based on a given number of
acceptable quality levels (AQLs) that are usually product specification, if this batch is acceptable.
applied to the various reinforcement materials.
For the purposes of this International Standard, two
main sampling methods are considered: by attri-
butes and by measurements (variables).
2 Normative references
3.2 elementary unit: The smallest normally com-
The following standards contain provisions which,
mercially available entity of a given product.
through reference in this text, constitute provisions
of thi ‘s International Standard. At the time of publi-
The description (form, dimensions, mass, etc.) of the
cation, the editions indicated were valid. All stan-
elementary unit will normally be defined in the
dards are subject to revision, and parties to
product specification. This unit may be supplied in
agreements based on this international Standard
one of several forms:
are encouraged to investigate the possibility of ap-
plying the most recent editions of the standards in-
- package (single yarn, folded or cabled yarn, rov-
dicated below. Members of IEC and IS0 maintain
ing, etc.);
registers of currently valid International Standards.
- roll (mat, woven fabric, veil, etc.):
IS0 472: 1988, Plastics -- Vocabulary.
- the smallest physical entity in a bulk product
IS0 28594 : 1989, Sampling procedures for inspection
(e.g. chopped fibres, milled fibres).
by attributes - Part I: Sampling plans indexed by
acceptable quality level (AQL) for lot-by-lot in-
NOTE 1 For a given product, the dimensions, mass or
spection.
volume of the elementary unit may change, as fabrication
techniques evolve, without necessarily causing any modi-
IS0 3534:1977, Statistics - Vocabulary and symbols.
fication in the product properties or the way in which
these properties vary throughout the elementary unit. As
IS0 3951:1989, Sampling procedures and charts for an example, while some years ago a bobbin of yarn had
a weight of 2 kg, the same item can now be supplied as
inspection by variables for percent nonconforming.
a 10 kg bobbin. In both cases, it is the bobbin which must
be considered as the elementary unit.
3 Definitions
3.3 case: The smallest conveniently handleable
unit (i.e. a carton or other container), which may
For the purposes of this International Standard, the
contain one or more elementary units of the same
definitions given in IS0 472, IS0 2859-1, IS0 3534
type and quality.
and IS0 3951 apply, plus the following definitions,

---------------------- Page: 9 ----------------------

SIST EN ISO 1886:1999
IS0 1886:1990(E)
3.4 batch (dispatched or received batch): A definite The number of nonconforming units is then com-
pared to the AQL given in the product specification
quantity of product made up of elementary units of
to determine whether the batch should be accepted
the same type and produced under conditions as-
or rejected (see example 1, subclause 7.1).
sumed to be constant.
A batch may constitute all or part of any particular
order.
3.8 inspection by measurements (variables): A
method of inspection that consists of measuring one
The cases making up a batch may be handleable
or more quantitative characteristics for each el-
individually, or several may by assembled together
ementary unit in a batch or in a sample taken from
on a pallet. All the cases constituting a batch are
the batch. Each quantitative characteristic must be
dispatched to, or intended for, a single customer and
measurable on a continuous scale. If more than one
accompanied by a single dispatch note.
characteristic is measured, it requires only one
characteristic to be unsatisfactory for the batch to
NOTE 2 In addition to the concept of a shipped batch,
be unacceptable.
the user of this International Standard should be aware
of the concept of a production batch. The latter concept
This method of inspection using measurements is
generally applies only when production takes place in
only applicable to characteristics that obey a normal
discontinuous runs. This is the case with carbon fibres, for
(i.e. Gaussian) distribution law.
instance, which must therefore be indentified by indicating
the production batch on each case. For continous pro-
For the purposes of this International Standard, the
duction (most textile-glass items, for example), the
measurement results obtained for each elementary
production-batch concept does not apply. Nevertheless,
each case should, in principle, be marked with a pro-
unit are used to calculate the mean for the sample
duction date. If, within a particular shipped batch, there
and the standard deviation. From these data, the
are widely separated production dates, manufacturer and
lower quality statistic & and/or the upper quality
customer may agree to consider sampling each pro-
statistic Qu are calculated, using the formulae given
duction run separately.
in example 2 (see 7.2.1), and QL and/or Qu then
compared with the acceptability constant k (see
table 2). A graph may also be plotted as an aid in
3.5 acceptability quality level (AQL): The maximum
assessing the quality of the batch sampled.
percent nonconforming (or the maximum number of
nonconformities per 100 units) that, for purposes of
Inspection by m eas urem ents may be applied to one
sampling inspection, can be considered satisfactory
or more propert ies with
as a process average.
- one limit, either lower or upper (see 7.2.1, ex-
The AQL will normally be defined, for each property,
ample 2);
in the product specification. It may not be the same
for all properties. In addition, properties subject to
-
two limits, either with different AQLs or with the
a lower limit and an upper limit may have a different
same AQL.
AQL for each limit.
In the case of two limits with different AQLs, batch
3.6 sample: A given number of elementary units
acceptance calculations are carried out first for one
which have been selected at random with a view to
limit, then for the other (see 7.2.2). In the case of two
performing a test either directly on these units or on
limits with the same AQL, the batch acceptance de-
one or more specimens taken from the units.
cision is based on a graphical treatment of the re-
sults (see 7.2.3, example 3).
NOTE 3 Where necessary, an additional sampling stage
may be carried out on these elementary units to give a
laboratory sample (e.g. a single piece of fabric or a given
3.9 operating characteristic curve: Each sampling
amount of chopped fibres). Instructions for the prep-
plan is characterized by an operating characteristic
aration of this laboratory sample will normally be given in
curve which indicates, for a given AQL, the prob-
the test method concerned.
ability Pa of a batch being accepted, depending on
the actual (i.e. measured) level p of nonconformities,
3.7 inspection by attributes (counting nonconform- in percent, in the batch.
ities): A method of inspection that, for every selected
The graph shown in figure1 is an example, taken
elementary unit, determines either
from IS0 3951, for a sample of 15 units and an AQL
of 1,5 % (the “s” method). If the product supplied
ence of one or more quali-
- the prese nce or abs
contains 4b/o nonconformities, about 62 % of the
tative cha racteristics ; or
batches submitted would be expected to be accept-
able using this sampling plan.
-
the conformance or not of a property with one or
more limits defined in the manufacturer ’s speci-
fication.
2

---------------------- Page: 10 ----------------------

SIST EN ISO 1886:1999
IS0 1886:1990(E)
on the actual standard deviation measured
based
on the sample.
90
m
a
4.2 Attributes or measurements?
s 80
iii
i? 70
CD?
When a batch is to be sampled, the method of in-
s- 60
OQ>
--
spection, i.e. by attributes or by measurements,
4
50
%$
must be decided first.
40
g8
6
30
Em
a) Inspection by attributes is simpler, especially as
Q)o
E!4 20
far as the decision, taken after sampling, as to
2s
10 the acceptability of the received batch is con-
n
cerned. This method of inspection can be used
14161820 22
w 2 4 6 8 10 12
with one or several qualitative parameters and
Percentage p of nonconformities
does not require any knowledge of the distrib-
in the batches submitted
ution law governing the results. It does require,
however, a greater number of elementary units
Figure 1 - Example of an operating characteristic
in the sample than does inspection by measure-
curve
ments.
b) Inspection by measurements cannot be used if
the law governing the distribution of the results
Operating characteristic curves of this kind allow an
is not a Gaussian one. However, inspection by
estimate to be made of the risk ,!? to the customer in
measurements requires, for a given AQL, a
accepting batches with a nonconformity level
smaller number of elementary units compared
greater than the AQL. Conversely, the risk oc to the
with inspection by attributes. Indeed, the use of
manufacturer of having a batch rejected which con-
the results themselves and their distribution
tains less nonconformities than the AQL can also be
provides much more information on the quality
estimated.
of the batch. Inspection by measurements is thus
particularly suitable for use with destructive
Manufacturer ’s and customer ’s risks can be as-
testing, complex test procedures or products for
sessed using the operating characteristic curve. For
which quality is a critical factor. When this in-
example, it can be seen in figure 1 that the horizon-
spection method is used in conjunction with
tal line corresponding to Pa = IO % cuts the oper-
quality-control charts, it is also a good way of
ating characterist.ic curve at a nonconformity level
detecting shifts in quality.
of 13,38 %. This means that, in 10 % of the cases,
the customer may have to accept with this sampling
The choice of sampling plan must thus take into
plan a batch containing in practice 13,38 % noncon-
consideration
formities.
-
the type of test which will be conducted and its
As for the manufacturer, there is a 5 % probability
cost;
of him having a batch rejected by a customer al-
thouqh it contains only I,09 % nonconformities.
.
-
the distribution of the measured values.
It should be noted that it is possible to use in-
4 Choice of sampling plan
spection by measurements for certain parameters
and inspe
...